Beyond Distributions and Primary Goods: Assessing Applications of Rawls in Information Science and Technology Literature Since 1990

Home , Primary goods

ADVANCES IN INFORMATION SCIENCE Beyond Distributions and Primary Goods: Assessing Applications of Rawls in Information Science and Technology Literature since 1990

Anna Lauren Hoffmann School of Information, University of California, Berkeley, 102 South Hall, Berkeley, CA 94702. E-mail: [email protected]

This article reviews the ways in which the work and of justice through which to assess various features of con- ideas of political philosopher John Rawls have been temporary social, political, and economic life. In particular, appropriated or applied by scholars of information sci- appealing to Rawls to address moral and political challenges ence, technology, and related areas since 1990. The article begins with an overview of Rawls’s work, paying of the development and use of advanced information and particular attention to its foundations and methods. communication technologies (ICTs) has scholarly prece- Subsequently, a 2-phase discussion of the literature is dent—notable efforts include Bell’s (1973) discussion of presented. The first phase reviews engagements with “post-industrial society,” Benkler’s (2006) liberal political Rawls from more than 150 scholarly articles from databases and journals dedicated to information and economic analyses of networked production, van Dijk’s technology ethics, information/technology studies, phi- (2005) account of the “digital divide,” and Sclove’s (1992) losophy of technology, and technology and human val- work on democracy and technological design. Rawls has ues. The second phase focuses on scholars who have also been employed extensively in conversations of moral- demonstrated a significant commitment to Rawlsian ity, ethics, and ICTs. This work represents an important con- theory in these areas. Overall, the review demonstrates that most significant applications of Rawls are oriented tribution to ethical analyses of ICTs by bringing together toward unequal distributions of informational goods and critically assessing the application of a particular—and exacerbated by advanced information and communica- particularly prominent—moral framework. tion technologies (ICTs). Critically, however, scholars in The article begins with an overview of Rawls’s theory of this area have overlooked the relevance of Rawls’s foundations for the sorts of complex networked relation- justice, paying particular attention to its foundational 1 ships afforded by those same advanced ICTs. Rather assumptions and methodological commitments. Next, a than representing a fatal gap, it opens up a new avenue two-phase discussion of the literature is presented. This first for the renewed consideration of Rawls—his ideas on phase attends to a wide range of engagements with Rawls the basic structure of society and the importance of based on more than 150 scholarly articles from databases self-respect are offered as 2 possible paths forward. and journals dedicated to considerations of information and technology ethics, information science, philosophy of information, philosophy of technology, technology and human Introduction values, and related areas. The second phase focuses on This article presents a comprehensive overview of how scholars whose bodies of work demonstrate a significant scholars of information and technology (broadly construed) 1Given that Rawls’s scholarship spans decades and thousands of have incorporated the ideas of political philosopher John pages, an exhaustive account of his work will not be possible here. In Rawls—among the most prominent political philosophers of the interest of efficiency, I have tried to focus on the most salient the 20th century. Rawls has long offered scholars from a aspects of his central theory—justice as fairness. Necessarily, certain range of disciplines a comprehensive and systematic vision tangential concepts and discussions are omitted. At the same time, however, some citations of Rawls in the literature make reference to concepts omitted from this initial discussion. In order to keep the literature Received August 26, 2015; revised January 20, 2016; accepted March review as complete as possible, I have—where necessary—incorporated 27, 2016 brief discussions of concepts that may not have been adequately addressed earlier. For a more thorough accounting of many of these con- VC 2016 ASIS&T Published online 21 December 2016 in Wiley Online cepts, please see the comparable review that first appeared in Hoffmann Library (wileyonlinelibrary.com). DOI: 10.1002/asi.23747 (2014) (doctoral dissertation).

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY, 68(7):1601–1618, 2017 commitment to the development of Rawlsian theory with (Rawls, 1980/1999a, p. 307). Put another way, moral con- regard to information and technology: Drahos (1996), van structivism adopts principles not because they are true, but den Hoven (1994, 1995, 1997), Brey (2000, 2007), Britz because they issue “from social procedures that are, in some (2004, 2008), and Duff (2006, 2011, 2012). Each author sense, suitable”—this is acceptable by all to whom they shares a set of core conceptions and commitments: They are apply (Darwall, Gibbard, & Railton, 1992, p. 139). The all committed to the idea that access to information should practical task before the constructivist, then, is to identify be a basic liberty; they each argue that information is vital to “what social procedures [are] suitable” (Darwall et al., p. the pursuit of individual and collective ends; and they focus 139). It is this task that Rawls sees before him in developing on the concept of information as a resource in the abstract, an account of justice. in particular by conceiving of information as a primary Rawls views the basic structure of contemporary liberal good. Combined with the most common uses of Rawls democratic societies as the primary problem for a theory of revealed by the broader review, these central concerns come justice (Rawls, 1971, p. 4). He gives two kinds of reasons for together to form what I will refer to as the “standard taking the basic structure as his primary subject. Under the account” of Rawls in informational and technological con- first kind of reason, he argues that principles of justice are texts. At the end of the article, I sketch the main ideas of the necessary for the regulation and preservation of just back- standard account and point toward some overlooked Rawl- ground conditions—what Rawls refers to as “background sian connections for future research and discussion. justice”—against which individuals set and pursue valued ends in line with their philosophical, moral, or religious John Rawls and Justice as Fairness beliefs, their sentiments and loyalties, as well as their fundamental values and projects, among other things (what Rawls Committed to the idea that there can be a reasonable, pub- refers to as a person’s comprehensive doctrine). Under the lic basis for argument on moral issues, Rawls’s work sets out second kind of reason, Rawls (2001) centers on the basic to develop a basis for reasoning about justice in contemporary 2 structure for its “profound and pervasive” influence on the liberal democratic societies. Initially, Rawls had hoped that life chances of citizens, given that it is integral to the structur- such a procedure might be useful for constructing a wide ing of individuals’ political, social, and economic possibilities range of ethical principles, but as his career progressed he (p. 55–56). Within the basic structure, Rawls intends his prin- refined and restricted the scope of its applicability, eventually ciples of justice as applying to the “basic institutions,” includ- limiting it to the construction of principles of justice (see, gen- ing a constitution and the system of government it defines, erally: O’Neill, 2003; Rawls, 1980/1999a, 1993). Perhaps his systems of property for regulating the use of goods, and eco- most notable achievement was the development of the nomic markets for distributing productive resources. “original position”—a constructive procedure for modeling Guided by his constructivist method, Rawls puts forward rational decision making under conditions that are fair (hence the “original position” as a suitable procedure for arriving at thenameofhistheory,justice as fairness). It is through this principles of justice suitable for regulating the basic struc- procedure that Rawls was able to revive and further the idea ture of society. Importantly, however, a great deal of the of a social contract as found in the work of Hobbes, Locke, “heavy lifting” of Rawls’s theory is done well before Rawls and—in different ways—Kant and Rousseau (Rawls, 1971, p. arrives at the original position. The device does not carry xviii). In addition, his work differs in important ways from justificatory force in and of itself, but instead works to oper- other strands of liberalism, especially classical economic lib- ationalize relevant features—referred to as “model- eralism and so-called neoliberalism, given that he does not conceptions”—of persons and society that he lays out in view liberalism as a fundamentally or narrowly economic advance.3 The first relevant model-conception is that of the doctrine tied to the rise of capitalism and market economic “well-ordered society”—that is, a society (i) in which every- thought. Rather, Rawls is situated in a liberal tradition that (i) one accepts (and knows others also accept) the same princi- is fundamentally concerned with ideals of toleration and the ples of justice and (ii) its basic structure is believed by all to promotion of basic freedoms of conscience and expression satisfy the principles of justice (Rawls, 1980/1999a, p. 309). and (ii) identifies the roots of liberalism in the 16th and 17th The second relevant model-conception—the “moral per- century Wars of Religion and the Reformation and their “at son”—describes members of a given society that view them- first, reluctant acceptance of the principle of toleration and selves, in their political and social relations, as free, equal, liberty of conscience” (Rawls, 2007, p. 11). and moral. For Rawls, persons are free if (i) they think they Methodologically, Rawls’s work represents a type of areentitledtomakeclaimsoncommoninstitutionsinorder moral constructivism. Whereas contemporary proponents of 3 moral constructivism differ, Rawls’s constructivism “holds It is important to note, however, that Rawls thinks societies and that moral objectivity is to be understood in terms of a suit- persons are more than these models. However, “if certain principles of justice would indeed be agreed to,” a constructivist account of justice ably constructed social point of view that all can accept” must connect definite principles with particular conceptions of the person and society (Rawls, 1980/1999a, p. 308). Other conceptions of per- 2The discussion of Rawls presented here has been greatly informed sons might be appropriate for reasoning about other moral principles, by Freeman (2007). For a more detailed version of this overview, please but insofar as we are concerned with principles of justice, we are, Rawls also see Hoffmann (2014, p. 27–68). thinks, concerned with the model-conceptions he advances.

1602 JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 DOI: 10.1002/asi to fulfill their goals and higher order interests and (ii) they stances of justice are conditions of moderate scarcity and the believe that they can revise or change their conception of necessity of social cooperation. They also include the fact of the good at any given time. Further, persons are equal inso- reasonable pluralism—an acceptance that a liberal demo- far as they recognize that everyone has an equal right to cratic society will inevitably need to account for a wide determine and evaluate the principles of justice that are to array of reasonable comprehensive doctrines that will be dif- govern the basic structure of society and they are moral ferent, often competing, and sometimes irreconcilable with insofar as they possess what Rawls calls the two moral one another. Finally, Rawls also stipulates that parties in the powers (discussed below) (Rawls, 2001). original position must employ public reasoning, that is, the Free and equal moral persons are also said to be both rea- reasons given for selecting particular principles over others sonable and rational. In crude terms, Rawls thinks a person must be justifiable to all whom the principles apply (in this has the capacity to be reasonable if they are prepared to case, citizens in a liberal democratic society). advance standards, principles, or reasons for action believed Rawls must also model the ideas of reasonableness and to be justifiable to (reasonable) others. On this account, rea- rationality in the original position. To model the idea of ration- sonableness falls somewhere between pure altruism (acting ality, Rawls charges parties in the original position with secur- solely on behalf of the common good) and mutual advantage ing sufficient means—or “primary goods”—for the exercise of (cooperating with others to advance one’s own ends) their two moral powers. Rawls’s index of primary goods is (Rawls, 1993, p. 50). By contrast, persons are rational when derived by ascertaining which things are generally necessary they exercise “powers of judgment and deliberation in seek- as both social conditions and all-purpose means not for advancing ends and interests peculiarly [their] own” (Rawls, 1993, ing persons’ interests generally under particular historical cir- p. 50). These ideas of reasonableness and rationality find cumstances, but for enabling persons to exercise their moral their expression in Rawls’s articulation of the idea of the powers specifically (Rawls, 1993, p. 314). The list of primary two moral powers: goods that Rawls adopts includes: basic rights and liberties; freedom of movement and free choice of occupation; powers First moral power: the ability of individuals to exercise an and prerogatives of offices and positions of responsibility in effective sense of justice, that is, a capacity to recognize and political and economic institutions; income and wealth; and act from justice’s demands (reasonableness) the social bases of self-respect. Among these primary goods, Second moral power: the capacity of individuals to possess the basic rights and liberties provide “for the adequate develop- and pursue a conception of the good, that is, to adopt some ment and full exercise of the two moral powers” (p. 45). For more or less complete set of valued ends and take up effec- promoting the first moral power (an effective sense of justice), tive means to those ends (rationality). Rawls (2001) provides persons with equal political liberties and freedom of thought (p. 45). The second moral power (the In addition, the complementary ideas of reasonableness capacity to adopt and pursue a conception of the good) is pro- and rationality (and their expression in the two moral moted through liberty of conscience and freedom of associa- powers) help Rawls in reconciling concepts of equality and tion (Rawls, 2001, p. 45). Rawls (1971) also affords special freedom that are often viewed by both critics and proponents consideration to the primary good of self-respect, calling it of liberalism as in tension with one another. The reasonable “perhaps the most important primary good” (p. 386). gives expression to equality insofar as it recognizes that all Reasonableness is modeled in the original position by sit- persons have an equal claim on matters of justice; the uating parties behind a “veil of ignorance” that deprives par- rational expresses the idea of the freedom of individuals to ties of information regarding “their place in society, their set, pursue, and revise individual ends. class position, or social status,” “their fortune in the distribu- These model-conceptions set conditions for the design of tion of natural talents and abilities,” and “their own distinctive the original position, which specifies “fair conditions under psychological dispositions and propensities” (Rawls, 1980/ which the representatives of citizens, viewed solely as free 1999a, p. 310). Behind the veil, parties are also denied partic- and equal persons, are to agree to the fair terms...whereby ular knowledge of their held beliefs and desired ends—in the basic structure is to be regulated” Rawls, 2001, p. 80). In Rawlsian terms, parties know they have comprehensive doc- this way, the original position incorporates “pure procedural trines, but do not know their doctrines’ specific contents. justice at the highest level” as “the fairness of the circum- Such restrictions, Rawls (1980/1999a) argues, are “required if stances under which agreement is reached transfers to the no one is to be advantaged or disadvantaged by natural con- principles of justice agreed to” (Rawls, 1980/1999a, p. 310) tingencies or social chance in the adoption of principles.” (p. (hence the name of Rawls’s theory, justice as fairness). 310). Overall, these informational parameters serve to situate Within the original position, Rawls imposes additional con- individuals as rationally autonomous and mutually disinter- straints on deliberation. One set of conditions—”the circum- ested, ensuring that deliberation between the parties is fair. stances of justice”—work to reflect (generally) our current Once inside the original position, parties deliberate and historical circumstances, given that if the principles selected select principles of justice for regulating the basic structure in the original position are to be useful, they must in of a well-ordered society. Loosely following the method of designed, in part, to be applicable to our actual conditions in ethics developed by Sidgwick, Rawls presents the parties the actual world (Rawls, 2001, p. 84). Among the circum- with different sets of principles of justice rooted in different

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 1603 DOI: 10.1002/asi philosophical traditions (including his own two principles of of justice. By putting forward an index of primary goods and justice) (Rawls, 2001, p. 83). It is important to note that the charging parties in the original position with securing their original position is designed in such a way so that the delib- fair share of such goods, Rawls is able to give expression to erating parties would most reasonably select his two princi- persons’ rationality. By excluding from consideration infor- ples over all others—that is, Rawls stacks the deck in his mation regarding persons’ specific place in society, Rawls is favor. This is not, however, a defect. Rather, it is a feature, able to give expression to persons’ capacities for reasonable- given that the important question for Rawls is not necessar- ness. Once inside the original position, Rawls further endeav- ily what principles are best, but what principles are best in ors to construct conditions under which reasoning about view of “one interpretation of the initial situation which best principles of justice can be construed as fair, employing vari- expresses the conditions that are widely thought reasonable ous informational restrictions—including the ideas of public to impose on the choice of principles yet which, at the same reason and the veil of ignorance. Finally, Rawls constructs time, leads to a conception that characterizes our considered theoriginalpositioninsuchawayastomaketheselectionof judgments” (Rawls, 1971, p. 105). This feature lays bare the his principles inevitable, reinforcing the importance of his fundamental importance of the assumptions Rawls builds foundational assumptions and model-conceptions.5,6 into the model-conceptions that inform the design of the original position. If one is critical of the principles Rawls’s Applications of Rawls to Information and parties adopt, it is not necessarily their reasoning inside the Technology: Overview original position that should be criticized, but, rather, the assumptions and constraints that Rawls builds into his This section offers a comprehensive overview of engage- model-conceptions in the first place. ments with Rawls based on more than 150 scholarly articles Ultimately, Rawls proposes two principles of justice: from databases and journals dedicated to considerations of information and technology ethics, information/technology 1. Each person has the same indefeasible claim to a fully studies, philosophy of information, philosophy of technol- adequate scheme of equal basic liberties, which scheme is ogy, technology and human values, and related areas from compatible with the same scheme of liberties for all; and 1990 to 2014.7 In addition to these sources, outside articles 2. Social and economic inequalities are to satisfy two conditions: First, they are to be attached to offices and posi- 5It should be noted that discussion of Rawls’s work on international tions open to all under conditions of fair equality of justice—especially as discussed in The Law of Peoples (1999b)—is opportunity, and second, they are to be to the greatest absent here. This is for two reasons: (i) It necessitates a subsequent dis- benefit of the least-advantaged members of society (the cussion that is, for the moment, outside the scope of this overview and (ii) within the review of applications of Rawls in the literature, only difference principle).4 Drahos (1996) makes any significant reference to Rawls’s work on international justice (where it is briefly discussed). This does, however, sug- Rawls intends these principles to specify a fair distribu- gest that Rawls’s international work may deserve more thorough or tion of primary goods in society. The first principle guaran- explicit attention from scholars interested in advancing Rawlsian theory tees individuals equal basic civil and political liberties as within the context of information and technology globally. (Many thanks to a reviewer for pointing this out.) well as the fair value of the political liberties (i.e., similarly 6Since the publication of A Theory of Justice in 1971, Rawls’s work able and motivated people should have the same prospects has spawned a great deal of subsequent discussion and secondary litera- for political participation). The second principle provides for ture, only a small amount of which is referenced later in this article. the effective actualization of the first: It ensures fair equality Though a full accounting of this secondary literature is outside the scope of any article-length piece, the following citations represent some nota- of social, professional, and economic opportunity and per- ble contributions and discussions: for prominent and wide-ranging dis- mits social, professional, and economic inequality only if cussions, see Daniels (1989) and Freeman (2003); for communtarian such inequalities work to the benefit of the least advantaged critiques, see MacIntyre (1981), Walzer (1984), Taylor (1989), and San- (the difference principle). Further, Rawls’s (2001) two prin- del (1998); for capabilities critiques, see Sen (2009) and Nussbaum ciples are lexically ordered: The first principle is necessarily (2004); for leftist critiques, see Cohen (2008) and Doppelt (1981); for disabilities critiques, see Nussbaum (2004) and Terzi (2010); for femi- preceding the first, whereas the first part of the second prin- nist critiques, see Jaggar (1983), Okin (1989), and Nussbaum (2003). ciple (fair equality of opportunity) is necessarily preceding 7Consulted databases include Library, Information Science, & Tech- the second part (the difference principle) (p. 43). The basic nology Abstracts (LISTA) and the digital library of the Association of civil and political liberties have special priority and cannot Computing and Machinery (ACM), both of which contain a wide range of publications relevant to ethics, information, and technology. Con- be compromised for second principle considerations. sulted journals include Ethics and Information Technology; Journal of As previously shown, much of the theoretical “heavy Information Ethics (JIE); International Review of Information Ethics lifting” of Rawls’s theory is done well before parties begin to (IRIE); Philosophy and Technology; Information, Communication, and select principles of justice within the original position, repre- Ethics (ICE); Science, Technology, and Human Values: Journal of the sented in his ideas of moral persons as free and equal, reason- Society for Social Studies of Science (ST&HV); Journal of the American Society for Information Science and Technology (JASIST); and First able and rational, and prepared to deliberate about principles Monday. Many of the more than 150 discovered articles are cited here, though some were omitted for the sake of space (especially if entries 4This articulation of the two principles of justice appears in Rawls were redundant or references to Rawls were misleading or particularly (2001, p. 42–43). fleeting)—hence the “reasonably comprehensive” qualification.

1604 JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 DOI: 10.1002/asi were consulted if they were frequently cited in works culled in order to better illustrate the alternative position. Examples from the primary search—though, admittedly, the number of of this approach include: capabilities analyses of morality outside articles is small. As the overview demonstrates, and robots as caretakers (Borenstein & Pearson, 2010; Val- references to Rawls in the areas of ethics, information, and lor, 2011); Coeckelbergh’s (2011) articulation of the capa- technology are common. They are not, however, scatter- bilities approach as a suitable foundation for a shot—rather, they tend to fall into one of four categories: (i) comprehensive “ethical-anthropological framework”; John- Rawls as a figurehead or representative of contemporary stone’s (2007) application of the capabilities approach for Western philosophy, in particular, social contract theory; (ii) normative analyses of technology generally; and Britz, Hoff- Rawls as an oppositional figure, that is, as representative of mann, Ponelis, Zimmer, and Lor’s (2013) discussion of an argument or position to be resisted; (iii) Rawls as an capabilities in an information-based rights framework. Out- authority for a particular normative concept, such as the idea side of capabilities discussions, Rawls is also cited as incom- of procedural justice or the priority or liberty; and (iv) Rawls patible with a feminist ethics of care (Capurro, 2008; as method, that is, as a method for reasoning about and arriv- Froelich, 2004). In a particularly clear example, Allen ing at moral rules or principles. Each of these four areas will (2013) rehearses both capabilities and feminist critiques of be discussed here. Rawls’s institutional focus in her discussion of frameworks for assessing disaster response efforts by nongovernmental Rawls as Figurehead organizations. In these cases, Rawls’s work is only indirectly Rawls is commonly cited as a figurehead for Western, criticized—it is simply presented to reinforce or justify the liberal philosophy or social contract doctrines. This type of choice of an alternative framework. reference often stands alone and serves to situate an author’s In other cases, Rawls’s work is not only used to punctuate work in relation to liberal philosophy generally without or illustrate an alternative approach, but is directly chal- explicitly engaging Rawls’s work (Carbo & Smith, 2008; lenged as well. Hands (2005), for example, criticizes Rawls’s Chang, 2011; Hongladarom, 2008; Kling, 1996; Laudon, model of public reasoning as an “idolatry of the state” that 1995). Garg and Camp (2012), for example, invoke Rawls mischaracterizes civil society as residing solely in the rela- as representative of “Western philosophies” generally (p. tionships between citizens and state institutions, and not also 10), whereas Hildebrandt (2011) and Duff (2005) cite him between citizens themselves (n.p.). Similarly, Noveck (2005) as a contemporary example of deontological ethical reason- challenges Rawls’s “idealization” of group deliberation ing. Clarke and Roache (2012) mention Rawls as typical of behind the veil of ignorance (n.p.). More broadly, Brothers contemporary liberal political philosophy, and Levy (2012) (1999) finds the liberal approach typified by Rawls wholly situates Rawls as an authority on the historical roots of liber- inadequate for addressing issues of social justice in a global alism. Similarly, Reed and Sanders (2008), Litschka and information society. Atkinson (2001) cites Rawls’s use of the Karmasin (2012), and Heeney (2012) recognize Rawls’s veil of ignorance as emblematic of a “neutral” liberal ideol- place in a liberal social contract tradition alongside Hobbes, ogy that limits our thinking about library services. Locke, and Rousseau (p. 1136). Hamlett (2003), Hands Oppositional applications have also been prevalent in (2005), and Brabham (2012) admit his work as representa- intercultural information ethics, where Rawls’s Western lib- tive of certain debates concerning deliberative and participa- eral commitments are taken as a signal of his ultimate incom- tory democracy. patibility with other cultural commitments. Ess (2007), for Rawls is also often invoked as a figure worth appealing example, argues that Rawls expects people to “[bracket their] to in future discussions. In these instances, a work describes diverse metaphysical beliefs” so that they might engage “with an ethical or political dilemma in relation to information [their] fellow citizens simply on the basis of what is politi- technology and cites Rawlsian justice as a possible solution. cally expedient” (p. 21). Wong’s (2012) discussion of ICTs In Wheeler’s (2003) critical discussion of the ACM Code of and the good life depends heavily on criticisms of Rawlsian Ethics, for example, Rawls’s political liberalism is cited as liberalism made by Michael Walzer (1984)—namely, that one possible way of systematically thinking through com- Rawls’s framework is not attentive to the social and cultural plex moral and political issues. Jain and Boehm (2005) sug- embeddedness of individuals. gest Rawls as potentially useful for attending to human values and software engineering. Finally, Kahn et al. (2010) Rawls as Authority on Normative Concepts cite uses of Rawls by developmental psychologists as a Outside of figurehead or oppositional uses, Rawls’s work future direction for philosophically grounded research in the is frequently cited for its clear articulation of particular nor- field of human-robot interaction. mative concepts. Sometimes, scholars of information and technology simply endorse Rawls’s normative prescriptions Rawls as Oppositional Figure wholesale. For example, Johnson and Miller (2002) invoke The inverse of figurehead approaches, oppositional uses Rawls’s two principles of justice as support for promoting reject Rawls’s framework as faulty or limiting. Here, equal opportunity in professional computing. Most exam- Rawls’s work is often contrasted against an already existing ples, however, are not so broad. Instead, specific dimensions critique—most commonly, Sen’s “capabilities approach”— of Rawls’s theory are discussed and endorsed as

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 1605 DOI: 10.1002/asi authoritative or normatively appropriate, namely: his articu- Additional considerations: Though the above features lation of procedural justice; his formulation of the difference represent the most commonly consulted dimensions of principle; his discussions of basic liberties and the priority Rawls’s work, other features have been invoked as well. of liberty; and the idea of an overlapping consensus. B€arwolff (2009) and Collste (2008), for example, cite Procedural justice: Computer scientists interested in Rawls’s conception of justice as generally preferable from a developing computational models of legal reasoning have normative perspective. Fallis and Whitcomb (2009) briefly turned to Rawls’s definition of procedural justice as a model cite Rawlsian rationality and the advancing particular ends. (Gordon, 1993; Leenes, 2001; Roth, Riveret, Rotolo, & Mathiesen (2013) has appealed to Rawls’s definition of a Governatori, 2007; Zeleznikow, Bellucci, Schild, & Mack- right as something that can be owed to a rights holder as a enzie, 2007). According to this work, procedural justice “is matter of justice. Dell and Venkatesh (2012) rely on Rawls’s concerned with making and implementing decisions accord- distinction between the reasonable and rational to show that ing to a fair process” (Zeleznikow et al., p. 238). Put another the exercise of reasonableness should be emphasized in way, just outcomes are the result of fair processes and are social design settings. In a discussion of reconciling conflict- not assessed independently of the procedures from which ing human values within value sensitive design and related they issue. A representative example is Lodder and Herc- design methodologies, Burmeister, Weckert, and William- zog’s (1995) “DiaLaw,” which builds on a definition of law son (2011) cite Rawls’s discussion of the tension between as “purely procedural” (in the Rawlsian sense) to create a the values of equality and efficiency. Eschenfelder, Glenn dialogical approach to modeling legal reasoning in comput- Howard, and Desai (2005) have focused on Rawls’s concep- ing environments. tion of civil disobedience in order to discuss practices of dig- Basic liberties and the priority of liberty: Rawls’s account ital rights management software circumvention. Palm of basic liberties and their priority has featured prominently (2009) has used the idea of just background conditions in in different works (Primeaux, 1998; Taebi, 2011). For exam- order to normatively assess workplace surveillance prac- ple, Laird (1993) cites Rawls’s assertion that, in democratic tices. Finally, Rawls’s idea of an overlapping consensus has societies, certain basic political liberties cannot be traded off also figured into discussions of ethics, information, and tech- for other goods (e.g., trading voting rights for money). Simi- nology—van de Poel and Zwart (2010) apply it in the con- larly, Franke (2012) invokes the lexical ordering of Rawls’s text of research and development (R&D) whereas two principles to argue against justifications of censorship Hausmanninger (2004) and Søraker (2006) draw on it in that appeal to second-principle considerations (such as eco- normative discussions of the Internet (though they arrive at nomic efficiency or property rights). Vaccaro and Madsen different conclusions). (2009) draw on van den Hoven and Rooksby’s (2008) interpretation of Rawls to articulate an “informational liberty” Methodological Applications (alternatively described as a “right to know”). In a different vein, Mattlage (2007) discusses Rawlsian basic liberties in In methodological applications, an author (or authors) the context of intellectual property rights, arguing that adopts a method or procedure from Rawls’s body of work Rawls’s “treatment of basic liberties recognizes the incom- for a particular applied context. For example, Collins, mensurability of certain primary (usually material) goods Miller, Spielman, and Wherry (1994) “use the reasoning that and the basic liberties of individuals” (p. 21). Rawls applied to general societal ethics in the specific con- Difference principle: Computer scientists often appeal to text of software fairness, modifying his ideas as needed to Rawls’s difference principle—alternatively referred to as the apply to computing” (p. 85). Similarly general approaches “maximin” principle (i.e., maximizing the minimum) in are taken by Lockhart (2001), Johnson and Miller (2002), these works—to argue for fair distributions of computing and Murphy (2012). More specifically, Powers (2003) uses resources within networks without appealing to utilitarian Rawls’s distinction between two concepts of rules to system- principles (Crowcroft & Oechslin, 1998; Dramitinos, atically identify norms and standards at play in a particular Stamoulis, & Courcoubetis, 2004; Ephrati, Zlotkin, & online community (see also Tavani, Grodzinsky, & Spinello, Rosenschein, 1994; Radunovic´ & Boudec, 2007). In a differ- 2003). Most methodological uses of Rawls, however, are not ent manner, scholars working in areas of information or digi- so broad. Rather, they tend to attend specifically to Rawls’s tal divides often appeal to the difference principle to show method of wide reflective equilibrium (discussed below) or how certain inequalities with regard to access to information the original position and its veil of ignorance. or technology are unjust (Bose, 2012; Britz & Ponelis, 2005; Wide reflective equilibrium: Rawls (2001) articulates the Duff, 2006; Fallis, 2004; Hodel-Widmer, 2006). Wolf and idea of “wide reflective equilibrium” to address conflicts Grodzinsky (2006) and Misra (2012) also cite the difference when “the implications of the judgments we render on one principle in this way, arguing that software engineers ought question [are] inconsistent or incongruent with those we to take into account the special considerations of the least render on other questions” (p. 30). In a state of wide reflective advantaged. In a unique application, Ottinger (2013) cites equilibrium, a person “has carefully considered alternative the difference principle as a potential alternative model conceptions of justice and the force of various arguments for for assessing the environmental risks of technological them” (Rawls, 2001, p. 31). Consequently, “we suppose this development. person’s general convictions, first principles, and particular

1606 JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 DOI: 10.1002/asi judgments are in line; ...the reflective equilibrium is wide, original position reasoning while generating their defense of given the wide-ranging reflection and possibly many changes “freedom zero” of the free software definition. of view that have preceded it” (Rawls, 2001, p. 31). In short, In less-detailed cases, other scholars have simply empha- a person is said to have reached a state of wide reflective sized the original position’s value as a framework for equilibrium when they have brought their general moral and addressing ethical issues generally. Floridi (1999), for exam- ethical principles more or less “in line” with their considered ple, admits the original position as a rare example of a judgments about particular cases or situations. For example, patient-oriented approach to moral reasoning in a typically the reasoning employed by parties in Rawls’s original posi- agent-centered Western philosophical tradition (see also Flo- tion represents one version of this method, given that parties ridi, 2006, 2010). Wallace (1999) holds up the veil of igno- move back and forth from considerations about specific pri- rance as reinforcing the value of anonymity for helping mary goods to broader moral theories (like utilitarianism or people make ethical decisions. Lercher (2008), in a discus- justice as fairness) that suggest or justify particular distribu- sion of moral dimensions of data-mining practices in health tions or societal arrangements. care and insurance industries, endorses the original position Various works have recognized the value of wide reflec- for limiting the role of risk taking in the selection of princi- tive equilibrium for reasoning about not only justice, but a ples of justice. In a different vein, Visala (1996) uses the wide range of ethical issues. van den Hoven (1997) has original position to defend a “universalizable” moral dis- argued that wide reflective equilibrium represents an appro- course and subsequently attempts to unite a Habermasian priate methodological approach for computer ethics, and model of communicative rationality with a Rawlsian model others have followed him in this assertion (Bynum, 2000; of rational decision making. Finally, Moor’s (1999) frame- Himma, 2008; Tavani, 2001). Elia (2009), in his discussion work of “just consequentialism” specifically relies on the of ICTs and transparency, describes the method as particu- impartiality of original position reasoning. For Moor, princi- larly valuable for “testing broad moral principles against ples of justice should guide in developing ethical policies for empirical data, situation difference, and intuitions or judg- using computer technology that protecting these fundamen- ments regarding specific cases in an ongoing and potentially tal goods (comparable to Rawls’s primary goods). To pre- revisionary way” (p. 147). van de Poel and Zwart (2010) serve this idea of justice, Moor develops a minimal and Doorn (2010) explicitly deploy the method in their impartiality test grounded in Gert’s (1998) “blindfold of attempts to morally assess particular R&D practices. Tidwell justice,” itself a modification of Rawls’s veil of ignorance. (1999) uses the method to discuss how the World Wide Web can support ethical debates between professional com- Applications of Rawls to Information and munities. Additionally, Clarke and Roache (2012) cite wide Technology: Major Proponents reflective equilibrium as a useful way of thinking through Beyond narrow or passing engagements with Rawls’s moral dimensions of human enhancement technologies. work, five scholars have emerged as notable proponents of Original position and the veil of ignorance: The most Rawlsian thought in informational and technological con- common methodological engagement with Rawls is through texts. In the mid-1990s, Peter Drahos began appealing to the device of the original position. Culnan and Regan (1995) Rawlsian thought in discussions of information justice. In rely on Rawls’s characterization of “mutually disinterested” particular, Drahos (1996) relied on Rawls’s two principles to persons in the original position to stress the importance of develop a systematic account of intellectual property rights preserving individual privacy in the face of massive political and just distributions of informational goods. Around the campaign mailing lists in the United States. Introna (2000) same time, Jeroen van den Hoven began advocating for the and Olivier (2002) invoke the original position as a way to value of Rawls’s work (van den Hoven, 1994). His later dis- ask what information privacy policies persons might choose cussion with Emma Rooksby (see van den Hoven & under ideal conditions. Vartiainen and Siponen (2010) rec- Rooksby, 2008) is, to date, perhaps the most comprehensive ommend the veil of ignorance as a useful tool for helping account of information as a primary good in the Rawlsian students think about moral dimensions of pirating software. sense. Philip Brey has stressed the value of Rawls for think- Johansson (2011) relies heavily on Brandt’s (1972) modifi- ing through moral dimensions of information and technol- cation of the veil of ignorance for assessing the morality of ogy. Johannes Britz (along with frequent coauthors Shana unmanned armed vehicles. Kaddu (2007) employs a modi- Ponelis and Peter Lor) has, on various occasions, relied on fied original position and veil of ignorance to advocate for a Rawls’s two principles of justice as a model for thinking free flow of information in society. LePoire (2005) suggests through just distributions of information goods on a global a version of the veil for resolving tensions between develop- scale. Alistair Duff has also promoted the use of Rawls for ment, risk, and uncertainty when developing technological studying issues of social justice in a global information soci- solutions to social problems. Cooke (2005) cites the original ety (or, as he often has it, a “post-industrial” society). Duff’s position as perhaps valuable for generating an ethical major work—A Normative Theory of the Information Soci- defense of lesbian, gay, bisexual, and transgender-specific ety (2012)—represents one of the most detailed interpreta- library and professional organizations, whereas Chopra and tions of Rawls to applied issues of information distribution Dexter (2009) develop a particularly robust application of broadly.

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 1607 DOI: 10.1002/asi Drahos and develops an alternative procedure where the representative parties are “peoples”—not individual rational persons In A Philosophy of Intellectual Property, Drahos (1996) (Rawls, 1999b). On Rawls’s account, because there is no addresses problems of information, intellectual property such thing as a global basic structure (i.e., a world govern- rights, and social justice through a Rawlsian lens. In order to ment), it does not make sense to design an original position account for information within a Rawlsian scheme of distrib- for arriving at principles of justice to govern a global basic utive justice, Drahos argues that information should be con- structure in the same way that his domestic theory is sidered a primary good. On his telling, persons need more designed to govern the basic structure of domestic institu- than equal rights and liberties in order to develop and pursue tions. Instead, parties in a global original position come plans and make decisions—they also need information (p. together to deliberate about terms for governing relations 175). Consequently, Drahos (1996) asserts that rational per- between states, so that the appropriate representatives in the sons would want more and not less information given that— original position will be “peoples” and not individuals. as with rights, liberties, opportunities, and resources These peoples, Rawls argues, would be most interested in (Rawls’s other primary goods)—information is indispensa- developing some basic rules and procedures for governing ble to pursuing one’s valued ends given that, “after all, the cooperation between states, as well as establishing some act of planning requires information” (p. 174). minimal institutions and agencies responsible for enforcing Drahos accounts for information as a Rawlsian primary these basic rules. In discussing global intellectual property good at two different levels: domestic and international. At agreements, Drahos emphasizes Rawls’s argument that peo- the domestic level, Drahos (1996) suggests a “freedom of ples in a global original position would likely adopt a posi- information” to sit alongside other basic liberties, like free- tion of noninterference, that is, that states should not dom of expression (p. 176). Here, the equal relevance and interfere in the inner workings of other states’ domestic import of information for individuals finds explicit protec- structures (so long as those structures meet certain minimal tion through Rawls’s lexically preceding first principle. At standards of decency). Drahos argues that robust protection- the same time, Drahos emphasizes the economic value of ist schemes of intellectual property rights would be barred information. He cites positive economics and Rawls’s differ- given that they would unduly interfere in domestic property ence principle as providing some justification for certain structures and undermine the autonomy of individual states productive inequalities in the distribution of information: (Drahos, 1996, p. 190). Inequalities in the distribution of information may be justified so long as they provide incentives to produce more van den Hoven information or goods that, ultimately, improve the well- being of the least advantaged (Drahos, 1996, p. 177). How- Jeroen van den Hoven has engaged Rawlsian theory on ever, Drahos (1996) is also quick to emphasize that certain multiple fronts. Broadly, he believes that Rawls’s systematic inequalities in the distribution of information could never be construction of principles of justice for well-ordered, demo- justified given that, under a Rawlsian scheme, second- cratic societies helped pave the way for considering issues principle considerations cannot take priority over first- of design in ethics (van den Hoven, 2010, p. 76). For van principle ones (p. 177–178). Instead, property rights in infor- den Hoven (2010), this “design turn” goes beyond offering mation should be viewed “as a tool to preserve political lib- simple applied ethical analyses to also address “the eco- erties and maximize access to, and the distribution of, nomic conditions, institutional and legal frameworks and primary goods such as information” (Drahos, 1996, p. 178). incentive structures that need to be realized if our applied To work this way, Drahos (1996) believes that parties in the analyses are to stand a chance in their implementation and original position would adopt a scheme of property rights in thus contribute to bringing about real and desirable moral information that severely limited monopoly rights (p. 179). changes in the real world” (p. 76). Whereas work in this Because monopoly rights in information limit the degree to vein is largely focused on institutional design, he believes it which information diffuses throughout the population, they simultaneously raises issues relevant to the design of socio- have the potential to generate unjust inequalities between technical systems and technological artifacts (van den information haves and have-nots by undermining the ability Hoven, 2008, p. 59; van den Hoven, 2010, p. 76). Addition- of the latter group to effectively exercise their first-principle ally, van den Hoven (1997) has argued that Rawls’s method rights and liberties. For this reason, parties in the original of wide reflective equilibrium (described earlier) is the most position, Drahos argues, would likely adopt a minimal (as appropriate methodology for considering ethical issues in opposed to maximal) scheme of intellectual property rights. the context of computers and other advanced ICTs (see also At the international level, Drahos adjusts his view of van den Hoven, 2008, p. 56–57). informational justice to accommodate the differences van den Hoven’s most thorough and pointed engage- between Rawls’s domestic and global theories of justice. He ments with Rawls position information as a primary good believes parties at the global level would also adopt a (1995; see also van den Hoven & Rooksby, 2008). As with severely limited scheme of intellectual property rights, Drahos earlier, van den Hoven argues that information quali- though their reasons for doing so would be different. At the fies as a primary good in the Rawlsian sense through its role global level, Rawls abandons his domestic original position in persons’ life planning (van den Hoven & Rooksby, 2008,

1608 JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 DOI: 10.1002/asi p. 382). Unlike Drahos, van den Hoven goes beyond simply Brey showing the instrumental value of information to also differ- Phillip Brey’s work is often concerned with clarifying entiate between different types of Rawlsian primary goods and synthesizing analyses of information and technology in his justification that information should be added to the issues across disparate disciplines or contexts. He has, for index. Recall that Rawls distinguishes between different example, proposed broader normative analyses of the Inter- types of primary goods: basic liberties, opportunities, and net and new media, seeking to develop an axiology of tech- income and wealth (all-purpose means). Citing Rawls’s nology—that is, a theory of values in technology—that assertion that the basic liberties are necessary background might help researchers, ethicists, and designers systemati- institutions for the exercise of the second moral power, van cally consider issues of power and value for both existing den Hoven and Rooksby (2008) argue that access to infor- and emergent technologies. He is also interested in develop- mation should be protected alongside established basic liber- ing methods and heuristics for uniting otherwise discon- ties like free speech or freedom of movement. Given the nected debates and issuing normative judgments in a lexical priority Rawls affords to protection of these basic lib- systematic and coherent manner. This interest led him to erties, van den Hoven is able to find and defend a basis develop a method of disclosive computer ethics that aims to within Rawls’s theory for information policies that promote systematize analyses of morally opaque features of techno- equitable access. logical systems (Brey, 2000, 2012). One of the strengths of Whereas access to information is amenable to Rawls’s this method is that it offers researchers and ethicists a clear index of basic liberties, information and ICTs also represent path for surfacing otherwise obscure or overlooked features opportunities and resources that can be, van den Hoven of technology that may have moral import. Once surfaced, thinks, further accommodated under Rawls’s second princi- morally problematic features can be subject to a theoretical ple of justice. In this way, the formal right to access informa- analysis according to established theoretical frameworks— tion already outlined is given substantive expression, given he often cites Rawls’s theory of justice as helping to morally that otherwise “a guarantee of equal liberty to seek informa- assess the justness of a technological artifact or system tion will not be sufficient in itself to ensure that all citizens through appeal to the ways it generates “an unequal distribu- have access to all relevant information” because certain tion of primary social goods according to Rawls’s theory of mechanisms are needed to ensure that people’s opportunities justice” (Brey, 2000, p. 13). to exercise their information liberties are roughly equal (van Unlike Drahos or van den Hoven, Brey does not attempt to den Hoven & Rooksby, 2008, p. 385). In describing access thoroughly or systematically account for information within a to information as a sort of opportunity, van den Hoven is Rawlsian framework. Instead, he is concerned with describing able to further accommodate information under the equal or accounting for informational and technological phenomena opportunities protections afforded by Rawls’s second princi- in ways that are congenial to Rawls’s system as Rawls himself ple of justice.van den Hoven’s work also attends to a num- describes it. For example, he often speaks of information or ber of critiques that have been leveled against Rawls’s work. access to technology as relevant additions to Rawls’s index of In particular, the work of Amartya Sen and Thomas Pogge, primary goods—but he is careful to justify how this might be as well as (though in a different way) communitarian according to Rawls’s own criteria (and not by critiquing the Michael Walzer have figured heavily into his thoughts on construction of Rawls’s index). Primary goods also feature information, technology, and justice. Thomas Pogge’s prominently in his conception of empowerment. In “The (1989) prominent socioeconomic critique has informed van Technological Construction of Social Power,” Brey (2008) den Hoven’s efforts to define information as a primary advances a theory of technology and power that is intended to good, and he readily admits the challenge of supplying peo- serve as a foundation for the development of a more-robust ple with effective means to the realization of basic liberties critical theory of technology. Here, individual agents are (including a right to access information). If some and not socially empowered for Brey (2008) if they are “free to deter- others have the means to exercise their basic liberties, the mine [their] own goals and choose [their] own actions, and ... overall worth of liberty remains unequal regardless of formal can exercise social power in relation to others to help attain protections. van den Hoven and Rooksby (2008) recognize these goals” (p. 6). Later, Brey (2008) relies on Rawlsian pri- this challenge, noting that “in high-technology information mary goods to further concretize this notion of empowerment societies, people [need to be] educated in the use of informa- as both “having the power to use one’s primary goods to tion technologies, and afforded access to information media one’s own ends” and “the successful acquisition of a relevant sufficient for them to be able to participate in their society’s share of primary social goods to more effectively further common life” (p. 389). More recently, van den Hoven has one’s ends” (p. 16). Consequently, Brey asserts that justice helped develop the use of Sen’s capabilities approach in the demands the adoption of design methods and regulations that context of technological design (Oosterlaken & van den promote the empowerment of individuals in the sense just Hoven, 2011). Finally, van den Hoven has also, at different described. Though he does not explicitly reference Rawls, it is points, cited criticisms of Rawls by Dworkin (choice- clear that he has Rawlsian justice and the difference principle sensitivity) and Garfinkel (positionality) and suggested their in mind in arguing that technology design should promote the implications for information and information technology. interests of the least advantaged (Brey, 2008, p. 21).

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 1609 DOI: 10.1002/asi In addition, Brey (2006, 2007) has also adopted the property holders to dictate the flow of their property whereas Rawlsian idea of comprehensive doctrines to describe the issues of access raise important questions about equality and value systems or ideologies held by individuals, groups, or the claims of users can make against the ownership and con- institutions when they make claims as to the goodness or trol rights of creators. badness of information and technology. In his axiology of Attending to this tension, Britz and various coauthors put new media, Brey relies on Rawls’s concept of comprehen- forward principles tailored specifically for addressing social sive doctrines to make sense of the lenses through which dif- justice and information, including: ferent individuals, groups, or institutions make value judgments about new media content and practices. For Brey 1. A fundamental right of access to information enshrined (2007), as for Rawls, comprehensive doctrines contain a alongside other basic liberties (Britz, 2008; Lipinski & thick conception of the good—detailed systems that define Britz, 2000; Lor & Britz, 2007; Ponelis & Britz, 2008); what one finds both intrinsically and instrumentally valua- 2. Following Rawls, all people should have access to the ble—and some more or less comprehensive set of accompa- same scheme of rights, including rights to communicate and to access information (Britz et al., 2006; Lor & nying beliefs and norms of conduct (p. 7). A normative Britz, 2007, 2012); axiology of new media, then, “utilizes a certain value system 3. Any protection of individual rights to control information or thick conception of the good to critique particular value should be justified by appeal to social utility or the com- implications of technology or culture” (p. 9). Examples of mon good and should not exacerbate the gap between explicit comprehensive doctrines that may be adopted in information rich and information poor or require unfair order to critique the value implications of new media include compensation (Britz & Ponelis, 2005; Britz et al., 2006; comprehensive religious doctrines (e.g., Christianity, Juda- Lipinski & Britz, 2000; Lor & Britz, 2007); and ism, or Islam), secularized consumer-oriented doctrines (i.e., 4. Inequalities in the distribution of information—or in the conceptions of the good promoted by marketers, advertising distribution of benefits and burdens in an informational agencies, and other similar institutions), or political doc- society—must be justified according to agreed-upon trines (e.g., liberalism, socialism, or communitarianism). norms or rules, like Rawls’s difference principle (Britz, 2004, p. 202; Britz, 2008). Brey (2006) similarly relies on the concept of comprehensive doctrines in exploring the social and cultural implica- Britz extends his concerns beyond these distributive prin- tions of the Internet in particular—he argues that making ciples to address important ideas of recognition, contribution, explicit the content and commitments of one’s comprehen- and participation in an information society. For example, he sive doctrine is integral to understanding a given appraisal argues that appropriately recognizing the equal worth and of the Internet or new media. dignity of all people should constrain any contemporary principles of justice (Britz, 2008, p. 1175). Additionally, Britz Britz (2008) is keen to the ways in which issues of justice might be Johannes Britz (2004, 2008) is largely concerned with obscured by existing structures and institutions. “These preset developing normative prescriptions for addressing the moral conditions make it difficult,” he writes, “to change or alter dimensions of a global information society. Early articula- society according to the moral imperatives set by justice” tions (see Lipinski & Britz, 2000) of this idea are rooted in (Britz, 2008, p. 1174). Britz (2008) argues—following Lotter€ interpretations of information and Rawlsian justice offered (2000)—that in-depth analyses are needed to examine injusti- by Drahos (1996) and van den Hoven (1995), though Britz ces obscured by embedded structures. later moves away from these interpretations of Rawls’s prin- In contrast to Rawls, Britz at times emphasizes ideas of ciples (Britz, 2008). Throughout these works, Britz attends desert that Rawls tends to de-emphasize. Britz’s (2008) to the tension between the control of information and access emphasis is clear in his definition of social justice: “to give to information. On the control side exist individual creators, a person or a group—in this case, all those who are part of publishers, and other intellectual property holders, as well as the global Information Society—what they deserve” (p. government agencies and affluent or “information-rich” 1174). Rawls (2001), on the other hand, is careful to distin- states; on the access side are users of information—citizens, guish between different kinds of desert and specify those students, other creators, and developing or “information- that are irrelevant for social justice and which are not (p. poor” states. In the context of global economic relations, 73). Of the types of desert Rawls outlines, one—moral this conflict plays out between “the right to own and con- desert—cannot be reconciled with his system of justice. trol” information asserted by affluent countries and transna- Because he admits that persons will have conflicting con- tional corporations and “the right of access to information” ceptions of the good, they will not be able to “agree on a claimed by developing nations often disadvantaged or fur- comprehensive doctrine to specify an idea of moral desert ther marginalized by global information policies (Britz, Lor, for political purposes” (Rawls, 2001, p. 73). Britz’s articula- & Bothma, 2006). This tension between control and access tion of desert is far less explicit, though he seems to rely on parallels the tension between liberty and equality found in a heavily moralized view of desert, as when he argues that Rawls’s work. Control, on Britz’s account, is concerned “everyone should get what they deserve—be it good or bad” with the freedom of information creators and intellectual (Britz, 2008, p. 1175).

1610 JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 DOI: 10.1002/asi Duff First Principle: Each person has an equal right to Type A information Alistair Duff has advanced a framework for addressing Second Principle: Inequalities in the social distribution of the problem of a just distribution of information in afflu- Type B information are permissible if and only if such ent Western democracies. Duff cites an informational inequalities: (a) maximize the informational resources of the dimension of digital divide debates often obscured by dis- worst-off group, (b) are subject to equality of access, (c) are cussion of devices or the Internet alone and he seeks to not extensive enough to cause class divisions (the Tawney “[postulate] a way of thinking about how to approximate proviso). (The distribution of Type C information, compris- a more ‘inclusive’ information society” (p. 605). To ing all other information, can be safely left to market forces.) address this divide, Duff (2011, 2012) appeals to Rawls (Duff, 2011, p. 609) (as well as the philosopher R.H. Tawney) to develop the “Rawls-Tawney theorem”—a normative framework mod- Formally speaking, the theorem emulates Rawls’s two eled directly on Rawls’s theory of justice (but limited in principles of justice, given that the first principle appears to certain ways by the fraternal ethos of Tawney’s work). As take priority over the second. Further, Duff’s second princi- with Drahos and van den Hoven earlier, Duff (2011) ple follows the multipart form of Rawls’s second principle appeals to the Rawlsian notion of primary goods, insisting because it accounts for both a version of the difference prin- that “in the postindustrial era ... information has grad- ciple and equality of opportunity. Finally, type C informa- uated into a primary good in Rawls’s sense” (p. 607). In tion is altogether excluded from consideration, whereas the this way, information can be treated as “an identifiable influence of Tawney can be seen in the second principle (the and separable good” of the type that can properly be sub- Tawney proviso). For Duff (2011), this proviso is important ject to a distributive scheme like the one articulated by because he believes that Rawls ultimately permits “very Rawls (Duff, 2011, p. 609). He further asserts that, within considerable socioeconomic inequalities” that, over time, the domain of information as a separable good, certain might undermine the achievement of social justice (p. 608). types of information are more important than others. Duff To account for this problem, Duff (2011) stipulates that not (2011) thinks that other information and technology theo- only should justifiable inequalities work to the benefit of the rists that have made use of Rawls have failed to “make least advantaged (as under Rawls’s difference principle), but clear whether [information] falls under [Rawls’] first or sec- they should also not be allowed to grow so large as to gener- ond principle of justice” and “when they do thus specify, they ate class divisions (p. 608). Beyond the emulation of Rawls’s principles in his Rawls-Tawney theorem, Duff tend to lump all information under either the first or the second, rather than pause to ask whether different types of infor- (2006) has also advanced a set of “neo-Rawlsian coordinates” for researchers to consider when considering the mation belong under separate principles” (p. 609). He does application of Rawls’s work to a range of issues in an infor- not give any particular reason as to why we must delineate mational, or “post-industrial,” society. between different types of information, but he is convinced that not doing so generates “weaknesses ... in most current Discussion and Possible Paths Forward neo-Rawlsian information society theory” (p. 609). To address this supposed weakness, Duff manufactures a The literature reviewed here shows that applications of hierarchical classification scheme that divides information Rawls’s work to information and technology issues have into three types: A, B, and C. Type A information includes been diverse, but, at times, converge upon similar themes. information relevant to citizenship, such as electoral infor- Outside of figurehead or oppositional citations, many uses mation and information regarding political and legal rights. are concerned with Rawls’s original position and—espe- Type B information is “within the orbit of social justice, cially the major proponents—focus on the just distribution [but] not so vital that it must be distributed exactly equally” of information as a discreet and commodifiable good. The and includes domestic and foreign news, as well as general most robust accounts in this area focus on information as a scientific, technical, and medical information (Duff, 2011, p. Rawlsian primary good; on these accounts, information 609). Finally, type C information includes “all other should count as a primary good because it is integral to information” (Duff, 2011, p. 609). Though he specifically rational life planning and for furthering human interests in cites soft news and entertainment as examples, type C would an informational—or post-industrial—society. The instru- presumably include all literary, artistic, and other cultural mental value of information finds expression in the articula- information not captured by the political, legal, scientific, or tion of rights of access to information, sometimes cast as a medical interests of types A and B. Finally, Duff deems this negative right (i.e., that one should not be hindered in the last category of information as unimportant in terms of pursuit of certain kinds of information) and other times a social justice. “In a world of scarce resources,” Duff (2011) positive right (i.e., that access to certain types of information writes, “the state cannot be burdened with the distribution of should be made accessible or facilitated by the state or other football scores, celebrity photo shots and the like” (p. 609). responsible agencies). The salient point here is that insofar Ultimately, Duff’s system is designed to articulate a just as Rawls has been substantively marshaled for use, it has distribution of information according to his hierarchical clas- largely been to (i) lean on his device of the original position sification scheme. His Rawls-Tawney theorem is as follows: and/or (ii) to discuss the distribution of information as a

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 1611 DOI: 10.1002/asi primary good. I will refer to this focus as the “standard to reinterpret contemporary informational or technological account” of Rawls in discussions of justice, information, and phenomena in terms congenial to Rawls’s system. technology. This approach is particularly evident in Drahos’s inter- The standard account’s focus on distributions is, in many pretation of intellectual property rights as well as various ways, unsurprising, given Rawls’s own emphasis on the dis- interpretations of information as a Rawlsian primary good tributive dimensions of social justice. Rawls (1971) himself (as in the works of van den Hoven, Brey, Britz, and Duff). describes his work as “[providing] in the first instance a The assumption here is that Rawls’s powerful practice- standard whereby the distributive aspects of the basic struc- independent conclusions are to be preserved and contextual ture of society are to be assessed” (p. 8). However, major considerations reinterpreted accordingly. As a consequence, proponents of the standard account have tended to adhere to the standard account of Rawls implicitly endorses the idea Rawls’s prescriptions without interrogating their founda- that principles of justice are to be worked out first and con- tions. They accept Rawls’s normative assumptions without textual considerations attended to second. It should be noted, offering particular consideration to the model-conceptions however, that such an argument fails to fully appreciate that underwrite justice as fairness and their relevance for the Rawls’s own motivations for developing an index of primary information or technological contexts within which they are goods. Rawls denies that primary goods account for things applied. Put another way, the standard account tends to deemed generally necessary for advancing persons’ interests approach Rawls’s two principles of justice as largely settled. under particular historical circumstances. In light of this, Indeed, it is the assumed “practice-independence”—that is, simply noting—as Drahos, van den Hoven, Brey, Britz, and its applicability despite contextual or practical concerns—of Duff all variously do—that information, because of its vital Rawls’s theory that seems, at times, to lend justice as fair- importance today, should be added to the index is perhaps ness its real normative force; through his sophisticated use shortsighted. Rawls did not arrive at his index of primary of original position reasoning, Rawls’s principles seem to goods by simply thinking long and hard about what avail- emerge from, as Sangiovanni (2008) puts it, a point of view able goods might best advance persons’ interests generally. that keeps us from constraining principles of justice by Consequently, it is not clear that we can simply add informa- social and political arrangements that may be, “at best, tion to the index just because we have thought long and hard merely the product of arbitrary historical contingency or, at about its importance as an all-purpose resource in an infor- worst, the result of past injustice itself” (p. 137). Specific mation society. Rather, one must keep in mind that Rawls’s features of practices or contexts, like information cultures or theory is designed to best promote the effective exercise of technological constraints, are secondary concerns—they persons’ two moral powers—that is, their capacities for a only become relevant once we seek to implement an inde- 8 sense of justice and to adopt and pursue a conception of the pendent conception of justice already worked out. 9 This is not to say that contextual considerations have good. been entirely absent. Proponents of the standard account In addition, the reviewed engagements have also tended have been careful to point out that Rawls’s ideas were not to foreground the authority of the original position and its formulated with advanced networked or information soci- veil of ignorance for Rawls, ignoring the device’s justifica- eties in mind. Rather, Rawls’s theory was worked out to tory foundations. Though it carries illustrative force, most of address problems of distribution against a backdrop of Rawls’s theoretical heavy lifting is done well before parties industrial social and economic practices typical of affluent begin to deliberate inside the original position. Importantly, Western democracies in the mid-20th century—hence its the model-conceptions that he develops and advances (free focus on the distribution of rights, opportunities, and mate- and equal persons, a well-ordered society) and their expres- rial wealth in the form of income and property. However, sion (two moral powers, primary goods) are not constructed the response to the limits of Rawls’s industrial perspective in the original position or behind the veil of ignorance. has not been to revisit the foundational assumptions of his Instead, they are simply laid out, their justification derived theory in order to assess their continuing relevance for soci- from Rawls’s constructivist method generally. This is not to eties dominated by sophisticated networks and technologies say that a rearticulation of Rawls’s model-conceptions or designed to support flows of information. Instead, the oppo- two moral powers would necessarily invalidate the conclu- site has happened: Rather than reinterpreting Rawls in light sions of standard account. Rather, it is only to note that inso- of new contextual considerations, the standard account seeks far as the principles Rawls arrives at fundamentally depend on the articulation of these model-conceptions in the first place—and whether or not they are capable of accounting 8It should be noted, however, that not all interpretations of Rawls for persons as producers and consumers of information has view him as ignoring certain contextual or socially contingent features. His introduction of the “circumstances of justice” into the original posi- 9At times, Drahos and van den Hoven are keen to Rawls’s justification, for example, can be read as accounting for some level of social tion for an index of primary goods in making explicit the ways in which and political context. In addition, proponents of the “practice- information supports the exercise the second moral power. However, dependent” interpretation of Rawls have endeavored to show how justice they still fall short of reconsidering Rawls’s model-conceptions of per- as fairness can be read as methodologically dependent on a notion of sons and society in light of the informational and technological practices social embeddedness (see, e.g., James, 2005; Ronzoni, 2007). they seek to address.

1612 JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 DOI: 10.1002/asi not been sufficiently addressed, suggesting a direction for decision-making power with relation to productive technolo- future work. gies in society, arguing that Rawls permits control of such In clarifying the motivations behind Rawls’s theory, I do technologies to concentrate into relatively few hands, lead- not mean to refute the idea that information should be consid- ingtounjustimbalancesinpower.Further,disabilitiescrit- ered a primary good subject to principles of distributive jus- ics have shown how normative standards of ability shape tice; indeed, attending to the justice of distributions of social and physical environments in ways that are biased informational goods is vital to the pursuit of justice today. against certain groups of people. For example, public build- However, this focus on distributions represents a particular (if ings that lack access for wheelchairs impose a normative necessary) limit: By reducing issues of social justice, informa- standard of mobility that excludes many otherwise capable tion, and technology to purely distributive terms, all we are persons. These debates share an attention to the ways that left to talk about are problems of distribution. But, as both the design of social, economic, and physical institutions critics and proponents of Rawls have pointed out, the focus assign roles and duties, structure decision-making power, on distributions tends to obscure or overlook other issues that and impose normative standards in ways that are relevant to are similarly—if not more—important to the realization of social justice but are not necessarily reducible to distributive social justice. As Iris Marion Young (2006) describes, “while terms. patterns of the distribution of resources, opportunities, and It should be noted, however, that some proponents of the income are very important issues of justice, theoretical focus standard account do, at times, account for problems of jus- on them tends to deflect attention from important aspects of tice in nondistributive terms. Brey, in particular, is commit- structural processes” (p. 91). She cites as examples the prob- ted to systematically exposing the values and biases lems of the social division of labor, structures of decision embedded within technological artifacts and systems—he is making power, and the elevation of certain judgments to nor- attuned to the ways in which not only the use, but also the mative standards—that is, processes in which “the attributes, design of technology can create and sustain relations that comportments, or ways of life that are ‘normal,’ in the sense empower some users and disempower others. Similarly, van of exhibited by ... dominant social segments, come to also den Hoven orients us toward the ways in which design is rel- have the connotation of being the ‘best’” (Young, 2006, p. evant to justice by making explicit the fact that Rawls’s prin- 95). Currently, the standard account inherits similar blind ciples of justice are intended to regulate not individual spots. It does not consider the structures or processes that pat- human action, but, rather, the design of the basic structure. tern distributions as also evaluable according to principles of Britz, too, is cognizant of nondistributive dimensions of jus- justice. Conceiving of information as a primary good, for tice, like recognition and contribution. Indeed, it is not diffi- example, cannot tell us much about the ways in which infor- cult to see how the design of technological systems mation is collected, framed, analyzed, presented, or pack- recognizes some (while rendering others invisible) or more aged—only about its dissemination. Consequently, the readily facilitates the contributions of certain people while structures and systems that allow people to access to informa- making participation more difficult for others. Whereas the tion—or the ways these systems may promote the distribu- work of Drahos and Duff limits us to thinking about purely tions of some types of information while at the same time distributive concerns with regard to information, Britz, hindering others—have overlooked in the standard account of Brey, and van den Hoven draw some paths forward for con- Rawls, information, and technology.10 sidering justice in information and technology outside of The importance of nondistributive dimensions of justice distributions. is well developed by critics of Rawls’s work. Many of these It should also be noted that insofar as these authors do critiques, however, have also gone overlooked by propo- pay attention to nondistributive issues, they often do not pro- nents of the standard account. For example, feminist cri- pose Rawlsian strategies for attending to them. For example, tiques of Rawls have long been aware of the unfair social in his discussions of privacy, van den Hoven appeals not to division of reproductive labor and the raising of children, a Rawls, but to the work of communitarian philosopher burden that—historically speaking—has disproportionately Michael Walzer (1984). Similarly, Britz and van den Hoven fallen on women. Leftist critics have pointed to problems of have, as of late, turned their attention to Sen’s capabilities approach, scaling back on their reliance on Rawlsian thought 10As one reviewer rightly pointed out, a focus on one thing (like dis- (Britz et al., 2013; Oosterlaken & van den Hoven, 2011). At tributive justice) does not automatically deny the importance of another first glance, this abandonment of Rawls when addressing thing (like nondistributive dimensions of justice). This is true. However, nondistributive dimensions of justice seems to point to limi- an overwhelming focus on one feature of a theory can, even inadver- tently, serve to obscure or distract from the importance of another fea- tations of the Rawlsian framework itself. The overview of ture. Given the dominance of primary goods-focused interpretations like, oppositional engagements with Rawls, for example, showed for example, van den Hoven and Rooksby (2008), I think this is the that many scholars resist Rawls in favor of the capabilities case here. Further—and as noted in the main text—many of the sharpest approach. This shift is, in some ways, unsurprising, given critiques of Rawlsian justice’s distributive focus (in particular, feminist that the capabilities approach is predicated (in part) on over- and disabilities critics) have also gone largely overlooked by scholars cited in the literature review. The relative inattention to these important coming perceived limitations of justice as fairness (Sen, discussions is another way the importance of nondistributive dimensions 1979, 1990, 2009). Famously, Sen (1979) argues that of justice can, I think, be said to be obscured. Rawls’s focus on primary goods might be appropriate if all

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 1613 DOI: 10.1002/asi people possessed roughly similar abilities to use such goods: tional insight for scholars interested in liberal justice, infor- “if people were basically very similar then an index of pri- mation, and technology. mary goods might be quite a good way of judging advantage” (p. 215). The problem, however, is that people Basic Structure as Infrastructure are not very similar. In fact, “people seem to have very dif- In spite of van den Hoven’s (2010) acknowledgment that ferent needs varying with health, longevity, climatic condi- Rawls’s focus on the basic structure equips us with powerful tions, location, work conditions, temperament, and even tools for thinking about ethics in the context of design— body size (affecting food and clothing requirements)” (Sen, scholars have largely overlooked the importance of the argu- 1979, p. 216). Against a primary goods metric, the capabil- ment from the basic structure. At the same time, scholars ities approach emphasizes the fulfillment of human needs from the area of science and technology studies have shown not in terms of subjective well-being or resources, but on the how the sociotechnical practices enabled by various infra- capabilities of individuals to convert available goods—like structures organize and make useful resources necessary for information—into substantive freedoms (Britz et al., 2013, the operation and upkeep of political, economic, and social p. 107–108; Robeyns & Brighouse, 2010). practices (Bowker & Star, 1999; Bowker, Baker, Millerand, Certainly, problems of informational and technological & Ribes, 2010; Ribes & Polk, 2012)—practices that make literacy—that is, the ability of different individuals to make up and sustain the very background conditions with which more or less effective use of available information and tech- Rawls is concerned. These discussions may offer new nology—seem better accounted for under a capabilities insight into how to best understand the relationships between model than one rooted in Rawlsian primary goods. I would institutions, individuals, and practices that make up Rawls’s suggest, however, that the abandonment of Rawlsian ideas basic structure. For example, the conventions of practice and in favor of the capabilities approach, however, has been the deployment of standards that allow infrastructures to somewhat premature. Whereas the capabilities approach connect a range of institutional and individual activities and provides valuable insight, it ultimately hinges on a narrowly become ubiquitous connect to Rawls’s notion of the basic teleological conception of technology as instrumental to structure’s pervasiveness. Additionally, the constraints gen- development and human flourishing. But the use of technol- erated by the proliferation of infrastructural standards, com- ogy is not simply instrumental—it is also intimately bound bined with the tangible disruptions caused by breakdowns in up with morality (see, e.g., Verbeek, 2009). Neither the infrastructure, capture Rawls’s idea of the profundity of the standard account of Rawls nor the capabilities approach are basic structure. Further, understanding that infrastructures able to account for the ways in which technology both medi- are built on installed bases—inheriting the capacities and ates our perception of morally relevant aspects of particular limitations of the bases they are built on (Star & Ruhleder, situations and actively shapes our responses to them. Given 1996)—lends empirical weight to Rawls’s claim that, that the problem of social justice is both a moral and practi- though any single transaction between individuals may be cal problem, technological mediation has implications for considered just, a great many transactions accumulating both conceiving of and achieving justice. Conceiving of over time may ultimately lead away from (and not toward) technology as merely instrumental is ultimately unsatisfac- justice. Consequently, Rawls’s discussions of justice and the tory. Any viable theory of social justice today ought to con- basic structure may offer normative insight for those inter- sider the how values embedded in the design of ested in the relationship between infrastructures and justice. technological artifacts and systems might actively promote—or hinder—social justice. Self-Respect: The Most Important Primary Good Against the standard account’s focus, the breadth of Although primary goods are discussed heavily in the Rawls’s work and its wide range of minor applications out- standard account, not all of Rawls primary goods receive the lined in the literature review suggest that there may be addi- same degree of consideration. In particular, the relationship tional resources available within Rawls’s work and its between information and self-respect—Rawls’s most impor- surrounding debates—a suggestion that has been echoed tant primary good—have not received sustained attention. elsewhere. Weber (2010), for example, has argued that Consequently, the ways in which sociotechnical relations scholars interested in intercultural information ethics have can influence the development of self-respect have been left ignored important developments and clarifications in the unexplored. Discussions of privacy, for example, highlight area of group and minority rights generated by debates over the ways in which technological advancements can threaten Rawls’s work (i.e., Kymlicka, 1989). In the remainder of individual dignity, integral to a sense of self-respect (Benn, this article, I briefly discuss two additional areas—the basic 1971; Bloustein, 1984). Moreover, the design of information structure and the value of self-respect—that might be fruit- technology routinely endorses certain abilities and strategies fully inform discussions of nondistributive dimensions of as normatively appropriate and others as marginal or inap- social justice, information, and technology. The following propriate. Questions regarding the values embedded in the discussions are, admittedly, brief sketches. However, they design of information systems (i.e., Friedman, Kahn, & are not intended to be complete arguments; I only mean to Borning, 2006) offer an entry point to examining how the point to two places where Rawls’s work may offer addi- promotion of certain values over others can similarly

1614 JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 DOI: 10.1002/asi promote the development of self-respect for some people the basic structure of society or the value of self-respect, and hinder it for others. For example, the routine promotion Rawls’s expansive body of work stands to offer scholars of of certain normative standards over others (e.g., in the design information and technology fresh lenses through which to of social networks that prioritize binary sex and make the examine pressing issues of social justice today. expression of non-normative sexes difficult) offers some and deprives others of the social bases of self-respect. Ultimately, References such a pattern of promotion and demotion of certain values can perpetuate already entrenched injustice by informally Allen, B.L. (2013). Justice as measure of nongovernmental organization deterring certain types of people from effectively engaging success in postdisaster community assistance. Science, Technology & with one another through technology or information systems, Human Values, 38, 224–249. Atkinson, R. (2001). Contingency and contradiction: The place(s) of the thwarting the equal opportunity that others—like van den library at the dawn of the new millennium. Journal of the American Hoven and Rooksby (2008)—would otherwise seek to pro- Society for Information Science and Technology, 52, 3–11. mote. A focus on the relationship between Rawls’s notion of B€arwolff, M. (2009). Discrimination, liberty, and innovation: Some the social bases of self-respect may offer some insight into thoughts on the invariable trade-offs of normative purposes and tech- issues of justice tied to the normative standards imposed by nical means in the internet. In Proceedings of the 2009 Workshop on Re-architecting the Internet (pp. 25–30). New York, NY: ACM. informational and technological systems. Bell, D. (1973). The coming of post-industrial society: A venture in social forecasting. New York: Basic Books. Conclusion Benkler, Y. (2006). The wealth of networks: How social production transforms markets and freedom. New Haven, CT: Yale University. Rawlsian analyses of information and technology are con- Benn, S.I. (1971). Privacy, freedom, and respect for persons. In J.R. ducted under the assumption that, despite their empowering Pennock & J.W. Chapman (Eds.), Nomos XIII: Privacy (pp. 1–27). potential, new ICTs “may also maintain, and even exacerbate, New York: Atherton. existing inequalities as they are grafted onto preexisting soci- Bloustein, E.J. (1984). Privacy as an aspect of human dignity: An answer to Dean Prosser. In F.D. Schoeman (Ed.), Philosophical oeconomic structures” (van den Hoven & Rooksby, 2008, p. dimensions of privacy (pp. 156–202). Cambridge, UK: Cambridge 377). The picture that emerges in these analyses is one of University Press. (Reprinted from New York University Law Review, information and technology as important Rawlsian primary 39, 962–1007, 1964) goods— vital instruments for the pursuit and achievement of Borenstein, J., & Pearson, Y. (2010). Robot caregivers: Harbingers of a wide range of human ends. Further, there exist plentiful expanded freedom for all? Ethics & Information Technology, 12, 277–288. depictions of Rawls’s original position reasoning, his veil of Bose, U. (2012). An ethical framework in information systems decision ignorance, and his two principles of justice. Normatively making using normative theories of business ethics. Ethics & Informa- speaking, these discussions are concerned with attending to tion Technology, 14, 17–26. the gap between information or technology “haves” and Bowker, G.C., Baker, K., Millerand, F., & Ribes, D. (2010). Toward “have-nots.” This focus on distributions is unsurprising, given information infrastructure studies: Ways of knowing in a networked Rawls’s own emphasis on distributive justice. environment. In J. Hunsinger, L. Klastrup, & M. Allen (Eds.), Interna- tional handbook of internet research (pp. 97–117). New York: Springer. However, much work on Rawls, information, and tech- Bowker, G.C., & Star, S.L. (1999). Sorting things out: Classification nology adopts Rawls’s distributive prescriptions without due and its consequences. Cambridge, MA: MIT Press. consideration to the assumptions that underwrite his theory. Brabham, D.C. (2012). The effectiveness of crowdsourcing public par- Scholars in this area have failed to attend to the relevance of ticipation in a planning context. First Monday, 17, n.p. Rawls’s foundational assumptions for the sorts of complex Brandt, R.B. (1972). Utilitarianism and the rules of war. Philosophy and Public Affairs, 1, 145–165. networked relationships afforded by advanced ICTs today. Brey, P. (2000). Disclosive computer ethics. ACM SIGCAS Computers This focus on distributions works to obscure or make and Society, 30, 10–16. invisible other dimensions that are equally important to the Brey, P. (2006). Evaluating the social and cultural implications of the realization of social justice. Similarly, conceiving of infor- internet. ACM SIGCAS Computers and Society, 36, 41–48. mation as a primary good cannot tell us much about the Brey, P. (2007). Theorizing the cultural quality of new media. Techne: Research in Philosophy and Technology, 11, n.p. ways in which information is collected, framed, analyzed, Brey, P. (2008). The technological construction of social power. Social presented, or packaged—only about the ways in which it is Epistemology: A Journal Knowledge, Culture, and Policy, 22, 71–95. disseminated. By reducing informational and technological Brey, P. (2012). Anticipating ethical issues in emerging IT. Ethics & goods to just (or mere) things to be distributed according to Information Technology, 14, 305–317. certain principles, standard discussions of Rawls, informa- Britz, J., Hoffmann, A., Ponelis, S., Zimmer, M., & Lor, P. (2013). On considering the application of Amartya Sen’s capability approach to an information- tion, and technology are unable to address the structures or based rights framework. Information Development, 29, 106–113. processes that pattern distributions. Consequently, the struc- Britz, J., Lor, P., & Bothma, T. (2006). Global capitalism and the fair tures and systems that allow for access to information—or distribution of information in the marketplace: A moral reflection the ways these systems may promote the distributions of from the perspective of the developing world. Journal of Information some types of information while at the same time hindering Ethics, 15, 60–69. Britz, J.J. (2004). To know or not to know: A moral reflection on infor- others—go overlooked from the standpoint of Rawlsian jus- mation poverty. Journal of Information Science, 30, 192–204. tice. By refocusing our attention on the foundations of jus- Britz, J.J. (2008). Making the global information society good: A social tice as fairness and on overlooked features like his focus on justice perspective on the ethical dimensions of the global information

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 1615 DOI: 10.1002/asi society. Journal of the American Society for Information Science and Duff, A.S. (2012). A normative theory of the information society. New Technology, 59, 1171–1183. York: Routledge. Britz, J.J., & Ponelis, S.R. (2005). Guidelines for fair distribution of Elia, J. (2009). Transparency rights, technology, and trust. Ethics & scholarly information. Mousaion, 23, 230–241. Information Technology, 11, 145–153. Brothers, R. (1999). Associative duties, institutional change, and agency: Ephrati, E., Zlotkin, G., & Rosenschein, J.S. (1994). Meet your destiny: The challenge of the global information society. ACM SIGCAS Com- A non-manipulable meeting scheduler. In Proceedings of the 1994 puters and Society, 29, 22–28. ACM Conference on Computer Supported Cooperative Work (pp. Burmeister, O.K., Weckert, J., & Williamson, K. (2011). Seniors extend 359–371). New York, NY: ACM. understanding of what constitutes universal values. Journal of Infor- Eschenfelder, K.R., Glenn Howard, R., & Desai, A.C. (2005). Who mation, Communication and Ethics in Society, 9, 238–252. posts DeCSS and why?: A content analysis of Web sites posting Bynum, T.W. (2000). The foundation of computer ethics. ACM SIG- DVD circumvention software. Journal of the American Society for CAS Computers and Society, 30, 6–13. Information Science and Technology, 56, 1405–1418. Capurro, R. (2008). Information technology as an ethical challenge. Ess, C. (2007). Cybernetic pluralism in an emerging global information and Ubiquity, 1. n.p. computing ethics. International Review of Information Ethics, 7, 1–31. Carbo, T., & Smith, M.M. (2008). Global information ethics: Intercul- Fallis, D. (2004). Social epistemology and the digital divide. In J. Weck- tural perspectives on past and future research. Journal of the Ameri- ert & Y. Al-Saggaf (Eds.), Selected papers from conference on com- can Society for Information Science and Technology, 59, 1111–1123. puters and philosophy, Vol. 37 (pp. 79–84). Darlinghurst, Australia: Chang, C.L. (2011). The significance of a suitable information ethical Australian Computer Society, Inc. Retrieved from http://www.acs.org. code: A case study of Chinese morality perspective. Journal of Infor- au/documents/public/crpit/CRPITV37Fallis.pdf mation Ethics, 20, 54–85. Fallis, D. & Whitcomb, D. (2009). Epistemic values and information Chopra, S., & Dexter, S. (2009). The freedoms of software and its ethi- management. The Information Society, 25, 175–189. cal uses. Ethics & Information Technology, 11, 287–297. Floridi, L. (1999). Information ethics: On the philosophical foundation Clarke, S., & Roache, R. (2012). Introducing transformative technolo- of computer ethics. Ethics & Information Technology, 1, 33–52. gies into democratic societies. Philosophy & Technology, 25, 27–42. Floridi, L. (2006). Information ethics, its nature and scope. ACM SIG- Coeckelbergh, M. (2011). Human development or human enhance- CAS Computers and Society, 36, 21–36. ment? A methodological reflection on capabilities and the evaluation Floridi, L. (2010). The philosophy of information as a conceptual frame- of information technologies. Ethics & Information Technology, 13, work. Knowledge, Technology & Policy, 23, 253–281. 81–92. Franke, U. (2012). Disconnecting digital networks: A moral appraisal. Cohen, G.A. (2008). Rescuing justice and equality. Cambridge, MA: International Review of Information Ethics, 18, 24–28. Harvard University Press. Freeman, S. (2003). The Cambridge companion to Rawls. Cambridge, Collins, W.R., Miller, K.W., Spielman, B.J., & Wherry, P. (1994). How UK: Cambridge University Press. good is good enough?: An ethical analysis of software construction Freeman, S. (2007). Rawls. Abingdon, UK: Routledge. and use. Communications of the ACM, 37, 81–91. Friedman, B., Kahn, P.H., Jr., & Borning, A. (2006). Value sensitive Collste, G. (2008). Global ICT-ethics: The case of privacy. Journal of design and information systems. In P. Zhang & D. Galletta (Eds.), Information, Communication and Ethics in Society, 6, 76–87. Human-computer interaction and management information systems: Cooke, J.C. (2005). Gay and lesbian librarians and the “need” for GLBT Foundations (pp. 348–372). Armonk, NY: M.E. Sharpe. library organizations. Ethical questions, professional challenges, and Froelich, T.J. (2004). Feminism and intercultural information ethics. personal dilemmas in and “out” of the workplace. Journal of Informa- International Review of Information Ethics, 2, 1–16. tion Ethics, 14, 32–49. Garg, V. & Camp, L.J. (2012). Gandhigiri in cyberspace. A novel approach Crowcroft, J. & Oechslin, P. (1998). Differentiated end-to-end Internet to information ethics. ACM SIGCAS Computers and Society, 42, 9–20. services using a weighted proportional fair sharing TCP. ACM SIG- Gert, B. (1998). Morality: Its nature and justification. Oxford, UK: COMM Computer Communication Review, 28, 53–69. Oxford University Press. Culnan, M.J., & Regan, P.M. (1995). Privacy issues and the creation of Gordon, T.F. (1993). The pleadings game: Formalizing procedural jus- campaign mailing lists. The Information Society, 11, 85–100. tice. In Proceedings of the Fourth International Conference on Artifi- Daniels, N. (1989). Reading Rawls: Critical studies on Rawls’ a theory cial Intelligence and Law (pp. 10–19). New York, NY: ACM. of justice (Vol. 229). Palo Alto, CA: Stanford University Press. Hamlett, P.W. (2003). Technology theory and deliberative democracy. Darwall, S., Gibbard, A., & Railton, P. (1992). Toward fin de siecle Science, Technology & Human Values, 28, 112–140. ethics: Some trends. The Philosophical Review, 101, 115–189. Hands, J. (2005). E–deliberation and local governance: The role of com- Dell, D.A., & Venkatesh, M. (2012). Social design’s implications for puter mediated communication in local democratic participation in the the IS field. In Proceedings of the 2012 iConference (pp. 346–353). United Kingdom. First Monday, 10. n.p. New York, NY: ACM. Hausmanninger, T. (2004). Controlling the net: Pragmatic actions or Doorn, N. (2010). A procedural approach to distributing responsibilities ethics needed? International Review of Information Ethics, 1, 1–10. in R&D networks. Poiesis & Praxis, 7, 169–188. Heeney, C. (2012). Breaching the contract? Privacy and the UK census. Doppelt, G. (1981). Rawls’ system of justice: A critique from the left. The Information Society, 28, 316–328. Nous, 15, 259–307. Hildebrandt, M. (2011). Who needs stories if you can get the data? ISPs Drahos, P. (1996). A philosophy of intellectual property. Aldershot, UK: in the era of big number crunching. Philosophy & Technology, 24, Dartmouth. 371–390. Dramitinos, M., Stamoulis, G.D., & Courcoubetis, C. (2004). Auction- Himma, K.E. (2008). The intercultural ethics agenda from the point of based resource reservation in 2.5/3G networks. Mobile Networks and view of a moral objectivist. Journal of Information, Communication Applications, 9, 557–566. and Ethics in Society, 6, 101–115. Duff, A.S. (2005). Social engineering in the information age. The Infor- Hodel-Widmer, T.B. (2006). Designing databases that enhance people’s mation Society, 21, 67–71. privacy without hindering organizations: Towards informational self- Duff, A.S. (2006). Neo-Rawlsian coordinates: Notes on a Theory of Jus- determination. Ethics & Information Technology, 8, 3–15. tice for the information age. International Review of Information Hoffmann, A.L. (2014). Google Books as infrastructure of injustice: Ethics, 6, 18–21. Towards a sociotechnical account of Rawlsian justice, information, Duff, A.S. (2011). The Rawls-Tawney theorem and the digital divide in and technology (doctoral dissertation). UWM Digital Commons postindustrial society. Journal of the American Society for Informa- Theses and Dissertations. Retrieved from http://dc.uwm.edu/cgi/view tion Science and Technology, 62, 604–612. content.cgi?article51535&context5etd

1616 JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 DOI: 10.1002/asi Hongladarom, S. (2008). Floridi and Spinoza on global information Mattlage, A. (2007). Intellectual property and coerced exchanges. Jour- ethics. Ethics & Information Technology, 10, 175–187. nal of Information Ethics, 16, 19–30. Introna, L.D. (2000). Workplace surveillance, privacy, and distributive Misra, H. (2012). E-governance and millennium development goals: justice. ACM SIGCAS Computers and Society, 30, 33–39. Sustainable development perspective in rural India. In Proceedings of Jaggar, A. (1983). Feminist politics and human nature. Lanham, MD: the Sixth International Conference on Theory and Practice of Elec- Rowman & Littlefield. tronic Governance (pp. 354–364). New York, NY: ACM. Jain, A., & Boehm, B. (2005). Developing a theory of value-based soft- Moor, J.H. (1999). Just consequentialism and computing. Ethics & ware engineering. In K. Sullivan (Ed.), Proceedings of the Seventh Information Technology, 1, 61–65. International Workshop on Economics-Driven Software Engineering Murphy, D.J. (2012). Are intellectual property rights compatible with Research (pp. 1–5). New York, NY: ACM. Rawlsian principles of justice? Ethics & Information Technology, 14, James, A. (2005). Constructing justice for existing practice: Rawls and 109–121. the status quo. Philosophy & Public Affairs, 33, 281–316. Noveck, B.S. (2005). A democracy of groups. First Monday, 10(11), Johansson, L. (2011). Is it morally right to use Unmanned Aerial n.p. Vehicles (UAVs) in war? Philosophy & Technology, 24, 279–291. Nussbaum, M.C. (2003). Rawls and feminism. In S. Freeman (Ed.), The Johnson, D.G., & Miller, K.W. (2002). Is diversity in computing a moral Cambridge companion to Rawls (pp. 488–520). Cambridge, UK: matter? SIGSCE. Bulletin, 34, 9–10. Cambridge University Johnstone, J. (2007). Technology as empowerment: A capability approach Nussbaum, M.C. (2004). The future of feminist liberalism. In A.R. to computer ethics. Ethics & Information Technology, 9, 73–87. Baehr (Ed.), Varieties of feminist liberalism (pp. 103–132). Lanham, Kaddu, S.B. (2007). Information ethics: A student’s perspective. Interna- MD: Rowman & Littlefield. tional Review of Information Ethics, 7, 2–5. Okin, S.M. (1989). Justice, gender, and the family. New York: Basic Kahn, P.H., Jr., Gill, B.T., Reichert, A.L., Kanda, T., Ishiguro, H., & Books. Ruckert, J.H. (2010). Validating interaction patterns in HRI. In Pro- O’Neill, O. (2003). Constructivism in Rawls and Kant. In S. Freeman ceedings of the Fifth ACM/IEEE International Conference on Human- (Ed.), The Cambridge companion to Rawls (pp. 347–367). Cambridge, Robot Interaction (pp. 183–184). Piscataway, NJ: IEEE Press. UK: Cambridge University Press. Kling, R. (1996). Beyond outlaws, hackers, and pirates: Ethical issues Olivier, M.S. (2002). Database privacy: Balancing confidentiality, integ- and computer science professionals. ACM SIGCAS Computers and rity and availability. ACM SIGKDD Explorations Newsletter, 4, 20– Society, 26, 5–15. in the work of information 27. Kymlicka, W. (1989). Liberalism, community, and culture. Oxford, UK: Oosterlaken, I., & van den Hoven, J. (2011). Editorial: ICT and the Oxford University Press. capability approach. Ethics & Information Technology, 13, 65–67. Laird, F.N. (1993). Participatory analysis, democracy, and technological Ottinger, G. (2013). Changing knowledge, local knowledge, and knowl- decision making. Science, Technology & Human Values, 18, 341–361. edge gaps: STS insights into procedural justice. Science, Technology Laudon, K.C. (1995). Ethical concepts and information technology. & Human Values, 38, 250–270. Communications of the ACM, 38, 33–39. Palm, E. (2009). Securing privacy at work: The importance of contex- Leenes, R.E. (2001). Burden of proof in dialogue games and Dutch civil tualized consent. Ethics & Information Technology, 11, 233–241. procedure. In Proceedings of the Eighth International Conference on Pogge, T.W. (1989). Realizing Rawls. Ithaca, NY: Cornell University Press. Artificial Intelligence and Law (pp. 109–118). New York, NY: ACM. Ponelis, S.R., & Britz, J.J. (2008). To talk or not to talk? From Telkom LePoire, D.J. (2005). Exploring ethical approaches to evaluate future to Hellkom: A critical reflection on the current telecommunication technology scenarios. Journal of Information, Communication and policy in South Africa from a social justice perspective. The Interna- Ethics in Society, 3, 143–150. tional Information & Library Review, 40, 219–225. Lercher, A. (2008). A social contract for health information. Journal of Powers, T.M. (2003). Real wrongs in virtual communities. Ethics & Information Ethics, 17, 35–45. Information Technology, 5, 191–198. Levy, N. (2012). Ecological engineering: Reshaping our environments to Primeaux, D. (1998). Using an alternative ethical paradigm for analysis: achieve our goals. Philosophy & Technology, 25, 589–604. An example regarding e-mail privacy issues. ACM SIGCAS Com- Lipinski, T.A., & Britz, J.J. (2000). Rethinking the ownership of infor- puters and Society, 28, 52–55. (June). mation in the 21st century: Ethical implications. Ethics & Information Radunovic´, B., & Boudec, J.Y.L. (2007). A unified framework for max- Technology, 2, 49–71. min and min-max fairness with applications. IEEE/ACM Transactions Litschka, M., & Karmasin, M. (2012). Ethical implications of the media- on Networking (TON), 15, 1073–1083. tization of organizations. Journal of Information, Communication, and Rawls, J. (1971). A theory of justice (revised ed.). Cambridge, MA: Ethics in Society, 10, 222–239. Belknap. Lockhart, C. (2001). Controversy in environmental policy decisions: Rawls, J. (1993). Political liberalism (expanded ed.). New York: Colum- Conflicting policy means or rival ends? Science, Technology & bia University Press. Human Values, 26, 259–277. Rawls, J. (1999a). Kantian constructivism in moral theory. In S. Free- Lodder, A.R., & Herczog, A. (1995). DiaLaw: A dialogical framework man (Ed.), Collected papers (pp. 303–358). Cambridge, MA: Harvard for modeling legal reasoning. In Proceedings of the Fifth International University Press. (Reprinted from “Kantian constructivism in moral Conference on Artificial Intelligence and Law (pp. 146–155). New theory”. The Journal of Philosophy, 77, 515–572, 1980) York, NY: ACM. Rawls, J. (1999b). The law of peoples: With, the idea of public reason Lor, P.J., & Britz, J.J. (2007). Is a knowledge society possible without revisited. Cambridge, MA: Harvard University Press. freedom of access to information? Journal of Information Science, 33, Rawls, J. (2001). Justice as fairness: A restatement. Cambridge, MA: 387–397. Belknap. Lor, P.J., & Britz, J.J. (2012). An ethical perspective on political- Rawls, J. (2007). Introduction: Remarks on political philosophy. In S. economic issues in the long-term preservation of digital heritage. Freeman (Ed.), Lectures on the history of political philosophy (pp. 1– Journal of the American Society for Information Science and Tech- 22). Cambridge, MA: Belknap. nology, 63, 2153–2164. Reed, G.M., & Sanders, J.W. (2008). The principle of distribution. Jour- Lotter,€ H.P.P. (2000). Christians and poverty. (Unpublished doctoral dis- nal of the American Society for Information Science and Technology, sertation.) Pretoria, South Africa: University of Pretoria. 59, 1134–1142. MacIntyre, A. (1981). After virtue. South Bend, IN: Notre Dame. Ribes, D., & Polk, J.B. (2012). Historical ontology and infrastructure. In Mathiesen, K. (2013). The human right to a public library. Journal of iConference ’12: Proceedings of the 2012 iConference (pp. 252–264). Information Ethics, 22, 60–79. Toronto, Ontario, Canada: New York, NY: ACM.

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 1617 DOI: 10.1002/asi Robeyns, I., & Brighouse, H. (2010). Introduction: Social primary goods van den Hoven, J. (1995). Equal access and social justice: Information and capabilities as metrics of justice. In H. Brighouse & I. Robeyns as a primary good. In ETHICOMP95: An International Conference on (Eds.), Measuring justice: Primary goods and capabilities (pp. 1–14). the Ethical Issues of Using Information Technology, Vol. 1. (pp. 1– Cambridge, UK: Cambridge University Press. 17). Leicester, UK: De Montfort University. Ronzoni, M. (2007). Two concepts of basic structure, and their relevance van den Hoven, J. (1997). Computer ethics and moral methodology. to global justice. Global Justice: Theory Practice Rhetoric, 1, 68–85. Metaphilosophy, 28, 234–248. Roth, B., Riveret, R., Rotolo, A., & Governatori, G. (2007). Strategic van den Hoven, J. (2008). Moral methodology and information technol- argumentation: A game theoretical investigation. In Proceedings of ogy. In K.E. Himma & H.T. Tavani (Eds.), The handbook of informa- the 11th International Conference on Artificial Intelligence and Law tion and computer ethics (pp. 49–67). Hoboken, NJ: Wiley. (pp. 81–90). New York, NY: ACM. van den Hoven, J. (2010). The use of normative theories in computer Sandel, M. (1998). Liberalism and the limits of justice (2nd ed.). Cam- ethics. In L. Floridi (Ed.), The Cambridge handbook of information bridge, UK: Cambridge University Press. and computer ethics (pp. 59–76). Cambridge, UK: Cambridge Univer- Sangiovanni, A. (2008). Justice and the priority of politics to morality. sity Press. Journal of Political Philosophy, 16, 137–164. van den Hoven, J., & Rooksby, E. (2008). Distributive justice and the Sclove, R.E. (1992). The nuts and bolts of democracy: Democratic value of information: A (broadly) Rawlsian approach. In J. van den theory and technological design. In L. Winner (Ed.), Democracy in a Hoven & J. Weckert (Eds.), Information technology and moral philos- technological society (pp. 139–157). New York: Springer. ophy (pp. 376–396). Cambridge, UK: Cambridge University Press. Sen, A. (1979). Equality of what? In S. McMurrin (Ed.), The Tanner van Dijk, J.A.G.M. (2005). The deepening divide: Inequality in the lectures on human values (pp. 197–220). Salt Lake City, UT: Univer- information society. Thousand Oaks, CA: Sage. sity of Utah Press. Vartiainen, T., & Siponen, M. (2010). On IS students’ intentions to use Sen, A. (1990). Justice: Means versus freedoms. Philosophy & Public theories of ethics in resolving moral conflicts. Journal of Information Affairs, 19, 111–121. Systems Education, 21(1), 33–42. Sen, A. (2009). The idea of justice. Cambridge, MA: Belknap. Verbeek, P.P. (2009). Moralizing technology: On the morality of techno- Søraker, J.H. (2006). The role of pragmatic arguments in computer logical artifacts and their design. In D. Kaplan (Ed.), Readings in the ethics. Ethics & Information Technology, 8, 121–130. philosophy of technology (2nd ed., pp. 226–243). Lanham, MD: Row- Star, S.L., & Ruhleder, K. (1996). Steps toward an ecology of infra- man & Littlefield. structure: Design and access for large information spaces. Information Visala, S. (1996). Interests and rationality of information systems devel- Systems Research, 7, 111–134. opment. ACM SIGCAS Computers and Society, 26, 19–22. Taebi, B. (2011). The morally desirable option for nuclear power pro- Wallace, K.A. (1999). Anonymity. Ethics & Information Technology, 1, duction. Philosophy & Technology, 24, 169–192. 21–31. Tavani, H.T. (2001). The state of computer ethics as a philosophical Walzer, M. (1984). Spheres of justice: A defense of pluralism and field of inquiry: Some contemporary perspectives, future projections, equality. New York: Basic Books. and current resources. Ethics & Information Technology, 3, 97–108. Tavani, H.T., Grodzinsky, F.S., & Spinello, R.A. (2003). Computer Weber, K. (2010). Information ethics in a different voice, or: Back to ethics in the post- September 11 world. Ethics & Information Tech- the drawing board of intercultural information ethics. International nology, 5, 181–182. Review of Information Ethics, 13, 6–11. Taylor, C. (1989). Sources of the self: The making of modern identity. Wheeler, S.L. (2003). An analysis of the association for computing Cambridge, UK: Cambridge University Press. machinery (ACM) code of ethics. ACM SIGCAS Computers and Terzi, L. (2010). What metric of justice for disabled people? Capability Society, 33, 2. and disability. In H. Brighouse & I. Robeyns (Eds.), Measuring jus- Wolf, M.J., & Grodzinsky, F.S. (2006). Good/fast/cheap: Contexts, rela- tice: Primary goods and capabilities (pp. 150–173). Cambridge, UK: tionships and professional responsibility during software development. Cambridge University Press. In Proceedings of the 2006 ACM Symposium on Applied Computing Tidwell, A. (1999). The virtual agora: Online ethical dialogues and pro- (pp. 261–266). New York, NY: ACM. fessional communities. First Monday, 4. n.p. Wong, P.H. (2012). A Walzerian approach to ICTs and the good life. Vaccaro, A., & Madsen, P. (2009). Corporate dynamic transparency: The Journal of Information, Communication and Ethics in Society, 10, new ICT-driven ethics? Ethics & Information Technology, 11, 113–122. 19–35. Vallor, S. (2011). Carebots and caregivers: Sustaining the ethical ideal of Young, I.M. (2006). Taking the basic structure seriously. Perspectives care in the twenty-first century. Philosophy & Technology, 24, 251–268. on Politics, 4, 91–97. van de Poel, I., & Zwart, S.D. (2010). Reflective equilibrium in R & D Zeleznikow, J., Bellucci, E., Schild, U.J., & Mackenzie, G. (2007). Bar- networks. Science, Technology & Human Values, 35, 174–199. gaining in the shadow of the law-using utility functions to support van den Hoven, J. (1994). Towards ethical principles for designing legal negotiation. In Proceedings of the 11th International Conference politico-administrative information systems. Informatization in the on Artificial Intelligence and Law (pp. 237–246). New York, NY: Public Sector, 3, 353–373. ACM.

1618 JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—July 2017 DOI: 10.1002/asi